Yubo Wang (王钰博)

I’m a CS Ph.D. student at TIGER-Lab in University of Waterloo, advised by Prof. Wenhu Chen. Previously, I obtained my Master’s degree in Computer Science from UC Davis. I received my Bachelor’s degree in Computer Science from Chu Kochen Honors College, Zhejiang University. From 2021 to 2023, I worked as a full-time NLP Algorithm Engineer at Tencent.

My research focuses on General Artificial Intelligence and its applications. I am particularly interested in understanding the behavior of LLMs, multimodal reasoning, and their evaluation.

News

Jul 08, 2025	🎉 Critique Fine-Tuning and ScholarCopilot are accepted to COLM 2025
Apr 13, 2025	I arrived at the University of Waterloo and started my Ph.D. journey!
Sep 27, 2024	🎉 MMLU-Pro is accepted to Neurips 2024

Selected Publications

Arxiv

Unleashing the Reasoning Potential of Pre-trained LLMs by Critique Fine-Tuning on One Problem

Yubo Wang, Ping Nie, Kai Zou, and 2 more authors

Jun 2025

arXiv Bib Website

@article{wang2025unleashing,
  title = {Unleashing the Reasoning Potential of Pre-trained LLMs by Critique Fine-Tuning on One Problem},
  author = {Wang, Yubo and Nie, Ping and Zou, Kai and Wu, Lijun and Chen, Wenhu},
  booktitle = {arxiv preprint},
  month = jun,
  year = {2025},
  github = {TIGER-AI-Lab/One-Shot-CFT},
  huggingface = {https://huggingface.co/collections/TIGER-Lab/one-shot-cft-683fbb4d2bcf698dbea8fb21},
  twitter = {https://x.com/YuboWang726/status/1930454445613088986},
  selected = true,
  num_co_first_author = {1},
}

COLM 2025

ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations

Yubo Wang, Xueguang Ma, Ping Nie, and 7 more authors

Apr 2025

arXiv Bib Website

@article{wang2025scholarcopilot,
  title = {ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations},
  author = {Wang, Yubo and Ma, Xueguang and Nie, Ping and Zeng, Huaye and Lyu, Zhiheng and Zhang, Yuxuan and Schneider, Benjamin and Lu, Yi and Yue, Xiang and Chen, Wenhu},
  booktitle = {2nd Conference on Language Modeling},
  month = apr,
  year = {2025},
  github = {TIGER-AI-Lab/ScholarCopilot},
  huggingface = {https://huggingface.co/collections/TIGER-Lab/scholarcopilot-6756f940df88114b24232074},
  twitter = {https://x.com/WenhuChen/status/1907861046833885397},
  selected = true,
  num_co_first_author = {1},
}

COLM 2025

Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate

Yubo Wang, Xiang Yue, and Wenhu Chen

Jan 2025

arXiv Bib Website

@article{wang2025critique,
  title = {Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate},
  author = {Wang, Yubo and Yue, Xiang and Chen, Wenhu},
  booktitle = {2nd Conference on Language Modeling},
  month = jan,
  year = {2025},
  github = {TIGER-AI-Lab/CritiqueFineTuning},
  huggingface = {https://huggingface.co/collections/TIGER-Lab/critiquefinetuning-679b25e1528e75180f55e5c4},
  twitter = {https://x.com/WenhuChen/status/1885060597500567562},
  selected = true,
  num_co_first_author = {1},
}

Neurips 2024

MMLU-Pro: A More Robust and Challenging Multi-task Language Understanding Benchmark

Yubo Wang, Xueguang Ma, Ge Zhang, and 8 more authors

Jun 2024

arXiv Bib Website

@article{wang2024mmlu,
  title = {MMLU-Pro: A More Robust and Challenging Multi-task Language Understanding Benchmark},
  author = {Wang, Yubo and Ma, Xueguang and Zhang, Ge and Ni, Yuansheng and Chandra, Abhranil and Guo, Shiguang and Ren, Weiming and Arulraj, Aaran and He, Xuan and Jiang, Ziyan and others},
  booktitle = {Proceedings of NeurIPS 2024},
  month = jun,
  year = {2024},
  github = {TIGER-AI-Lab/MMLU-Pro},
  huggingface = {https://huggingface.co/spaces/TIGER-Lab/MMLU-Pro},
  twitter = {https://x.com/YuboWang726/status/1839330905145610688},
  selected = true,
  num_co_first_author = {1},
}